Quote of the day

“I find economics increasingly satisfactory, and I think I am rather good at it.”– John Maynard Keynes
Showing posts with label statistics. Show all posts
Showing posts with label statistics. Show all posts

Wednesday, 12 February 2025

Time to scare you - "real" national debt figures

 

Statistical skulduggery is turning Britain into a joke

The national accounts are full of cons and tricks – it’s no wonder investors don’t take us seriously

Sir Keir Starmer
‘Sir Keir Starmer’s Government has doubled down on Britain’s high-debt, high-tax, low-growth doom loop’ Credit: Andy Rain/EPA

This column often focuses on the UK’s national accounts – with good reason. Britain’s public finances are in a parlous state, teetering on the brink of systemic meltdown.

In 1997, when Tony Blair took office, the economy was growing at a rate of 4.9pc a year and the national debt was 36pc of GDP – low by historic standards. Growth continued at 3-4pc per annum for the rest of that parliament – so a relatively small government debt pile remained manageable as a share of total output, allowing the state to borrow and spend more.

Keir Starmer’s incoming administration faced a different situation. Over the previous 14 years, the Tories had overseen a decade of tough fiscal rhetoric during which the national debt actually ballooned, followed by a wildly over-stringent Covid lockdown that weakened our public finances even more.

As a result, Labour entered government last July with Britain in a high-debt, high-tax, low-growth doom loop – with growth at a historically paltry 1.1pc and national debt near 100pc of GDP, a 60-year high.

But Starmer and Chancellor Rachel Reeves, leading a government far more ideologically Left-wing than the Blairites, doubled down. Last October’s Budget raised taxes by another £40bn a year so Reeves could jack up spending even more, not least by awarding sweetheart pay-deals to Labour’s public-sector union paymasters.

This strategy has proved disastrous, as some of us warned from the outset. Government borrowing costs have soared, with the 10-year gilt yield up from 3.7pc in mid-September to 4.5pc now – consistently way higher than during the “mini-Budget crisis” of October 2022 which saw Liz Truss drummed out of No 10.

The pension funds and insurance companies that lend governments serious money are spooked because, far from picking up under Labour, growth has slumped with Britain now probably in recession. Investors judge that a faltering, heavily over-taxed UK economy will struggle to raise the revenues ministers hope for, while Labour lacks the political grit to rein spending in.

The market consensus is Starmer’s Government will keep borrowing and spending even more, raising tax rates ever higher in a desperate bid to plug the gap, driving the economy ever deeper into the doldrums, making our fiscal position even worse.

That’s why gilt yields remain stubbornly high, forcing the Government to spend evermore billions on debt interest each month, despite the Bank of England repeatedly cutting its base rate in a bid to lower economy-wide borrowing costs. The UK’s national debt looks set to soar above 100pc of GDP – which will unnerve financial markets even more.

Yet, as tough as the fiscal outlook appears, the underlying reality is worse. Britain’s headline national debt figure is actually a serious under-estimate of the Government’s true debt burden.

Official public sector net debt (PSND) is £2,674bn on the latest 2023/24 data ­– equivalent to 98pc of GDP. That’s sharply up from 80pc of national output prior to lockdown and less than 40pc just before the 2008 global financial crisis.

This figure, though, is way lower than it should be if you include additional liabilities under three headings: the Bank of England’s asset purchase facility (APF); contractual debts accrued under the private finance initiative (PFI); and, the really big one, mammoth state obligations represented by the still very generous, inflation-proofed pensions enjoyed by millions of public sector workers.

As a result of the APF, the Office for National Statistics (ONS) makes a huge multi-billion-pound deduction from official national debt attributed to “The Bank of England” in our national accounts. This, in my view, is entirely unjustified.

In reality, the Bank is sitting on big losses as a result of its quantitative easing (QE) programme, under which the central bank bought hundreds of billions of pounds of mainly sovereign debt in a bid to boost the economy over the last decade or so. The value of those assets has since fallen. Yet instead of adding those losses to the national debt, the ONS – in a strange metaphysical twist – subtracts them from the UK’s headline debt figure.

“As well as being unjustified, this deduction is dangerous in that it normalises the routine understating of the nation’s indebtedness,” says Bob Lyddon of Lyddon Consulting, a highly-respected economic consultancy specialising in the scrutiny of bank balance sheets. “Such creative accounting leads only to one place: the invention of illusory headroom for further public sector borrowing”. If APF liabilities are added, the headline PSND figure, rises by £179bn to £2,853bn – from 98pc to 105pc of GDP.

To that should also be added the huge liabilities that are still outstanding under hundreds of PFI contracts, a trend which began under John Major’s Conservatives before accelerating sharply under Blair. PFI was used to disingenuously keep public investment off the state balance sheet by relying on private capital instead, with investors guaranteed huge, taxpayer-backed returns for years to come.

Often what was delivered under these contracts were extremely shoddy schools, hospitals and other public infrastructure. Add in £94bn of PFI liabilities, all of which must be legally met by the state over the coming years, and our national debt climbs further to £2,947bn, or 108pc of GDP.

Then there is the often-ignored bill for the very large, index-linked pensions due to civil servants, NHS staff, teachers and some other state workers – paid for not out of invested funds, but current and future tax receipts. That bill, officially estimated at £1,442bn but according to many experts much higher, takes the national debt to £4,389bn – an astonishing 161pc of GDP, two thirds above the headline figure.

The UK’s national accounts are a morass of statistical cons and tricks designed to make our national debt look smaller than it is – a culture Chancellor Reeves has embraced.

It’s no way to run a serious country. No wonder serious financial analysts, and the bond traders they advise, increasingly view Britain as a joke.

Thursday, 16 January 2025

The problem of inaccurate data coming from the ONS

 

Andrew Bailey: our statistical shortcomings have become a “substantial problem”

Britain is flying blind

Modern states rely on accurate data to guide policymaking. But Britain’s statistics authority is increasingly unfit for purpose. Why, and what can be done? Simon Wilson reports

What is the ONS? 

The Office for National Statistics (ONS) is the UK government’s statistics authority, charged with “collecting, analysing and disseminating statistics about the UK’s economy, society and population” – chiefly for use by government and policymakers. The ONS is the executive office of the UK Statistics Authority, with headquarters in Newport, South Wales, and since 2008 it has been a non-ministerial department that reports directly to parliament, rather than government. But there’s a growing feeling that some of its statistics are no longer fully reliable. In a Mansion House speech in November, the Bank of England’s governor, Andrew Bailey, said that unreliable labour-market statistics have become a “substantial problem” for the UK central bank in terms of policymaking. In particular, the Bank’s chief economist, Huw Pill, believes that ONS figures are probably understating employment growth and overstating the high rates of labour-force inactivity. In addition, in recent months the Office for Budget Responsibility and parliament’s Treasury Select Committee (plus several think tanks) have all flagged up fears about the reliability of ONS data – and the resulting adverse effects on policymaking. 

What’s causing the problem?

In terms of the labour market, a big part of the issue is falling response rates for the ONS’s Labour Force Survey (LFS). This vital survey involves a letter sent to households, and followed up with a phone or in-person interview. Whereas ten years ago response rates were about 50%,  they were falling even before the  pandemic, and then plunged during the various lockdowns. Since then, they’ve remained low, at just 17.3% in 2023  and are expected to be even lower in the past year. Ian Diamond, the ONS’s chief, says that, even after raising the incentives for participating in face-to-face interviews from £10 to £50, the public has become much more reticent. The reasons included “cautiousness around the sharing of personal information, declining trust in government and public institutions, a reluctance to have interviewers inside homes and increased challenges accessing secure/gated properties”. 

What are the implications?

The smaller sample means the data is less reliable and some economists think the LFS is now more likely to record people who are at home – thus overestimating the level of economic inactivity overall. The Resolution think tank estimates that the LFS may have “lost” up to 930,000 workers. And the ONS itself has downgraded the status of its employment data from “national statistic” (its gold-standard category) to “experimental”. That’s a problem, as it means policymakers are flying blind and “grand narratives rest on shaky numbers”, says The Economist. Why has Britain – alone among rich Western nations – suffered such high levels of economic inactivity post-pandemic? The question could scarcely be more important to the country’s future, yet it seems the phenomenon might well have been a “statistical artefact of increasingly dodgy data”. The LFS is “essentially useless”, says Xiaowei Xu of the Institute for Fiscal Studies. “It’s a huge problem.” 

Are improvements on the cards?

The ONS is working on a new “online-first” version of the survey, which will ask shorter questions, making it less hassle for participants, and boost participant numbers. The system is currently being trialled through introductory letters containing a QR code that can be scanned from a mobile phone. But last month Diamond admitted this probably won’t be ready to go live before 2027 – leaving the agency’s key stakeholders unimpressed. Meg Hillier, chair of the Treasury Select Committee, says the delay would rob policymakers of reliable data about the jobs market, making “some of the most consequential decisions taken by the Treasury and Bank of England challenging at best and misinformed at worst”.

Are there other problems?

Yes. When it comes to net inward migration, the ONS has been dramatically underestimating the numbers. Recently it revised its figure for 2023 from 740,000 to 906,000 (that’s almost another Oxford). And it upped its 2022 figures, too – from 606,000 to 872,000 (a second Stoke); or look at housing, where the ONS data is “not fit for purpose”, says Neal Hudson in the Financial Times. It relies too heavily on incomplete building-control data from one source, the National House Building Council, which undercounts the true number of new dwellings. The ONS’s numbers on production, trade and GDP are widely regarded as more reliable, says Andrew Sentance on CapX. Even so, the Living Costs and Food Survey, an important factor when it comes to assembling the GDP figures, now gets a similar low response rate to the LFS, making data less reliable and (even) more susceptible to later revision.

What is to be done?

The ONS deserves credit for being a generally trusted source of accurate data, says Sentance. But its ability to adapt its data collection and analysis techniques nimbly is “a vital part” of its remit, and it’s where the agency has sometimes been “slow and hesitant”. Problems with the LFS, for example, first emerged in 2021; a solution should be in place by now, not 2027. The ONS’s regulator, the Statistics Authority, should demand a 90-day action plan setting out how the agency proposes to become “action-oriented and responsive” to users’ concerns; and failing that, the Treasury should intervene and oversee improvements. As for response rates, the state needs to become far more assertive about compelling participation in nationally important data collection – in every area from the economy, to health, to the justice system, says The Economist. “Britain is a blind state.” Given the country’s deep-seated problems, that’s a self-inflicted wound it can scarcely afford. 


Wednesday, 25 October 2023

A good look at issues with gathering unemployment data

 


author-image
DAVID SMITH

Newport, we have a problem: how much can we trust ONS figures?

The Times
Share
Save

For those of us whose life is governed by the calendar of official statistical releases — sad but partly true in my case — something has been missing this week. The Office for National Statistics (ONS) has released some information on the labour market, but not the usual range of data.

We have had information on pay, which is now outstripping inflation but shows signs of decelerating, and on job vacancies, which are falling quite sharply. They are now below one million and have fallen by more than 250,000 over the past year. There was also limited information on employment, which appears to be falling. What was missing, however, was what used to be regarded as the centrepiece of these data releases: what is happening to unemployment, and to employment, on the ONS’s preferred measure, based on the labour force survey (LFS).

That information has been delayed until October 24, as was announced by the statistical agency in a surprise press release on Friday. That does not sound too dramatic. However, behind that week-long delay lies significant problems with what used to be regarded as a gold standard of statistical measurement.

It may surprise some readers to learn that the unemployment and employment statistics are based not on an actual count, but on what is claimed to be the largest household survey in the UK. Just as opinion polls of, say, 1,000 people can give a good indication of the state of the political parties, so the LFS, which covered more than 53,000 individuals in nearly 25,000 households in the latest quarter, can do the same for the labour market.

That at least is the idea, and it has worked well until now. But the survey has run into problems, some of which are a direct result of the pandemic, and some of which were emerging before it. A survey is only as good as its sample and the LFS is plagued by falling response rates and the difficulty of getting to all the types of people it wants to. Face-to-face interviews, which became impossible during Covid, gave way to phone polling, but this suffers from the problem that many people no longer use landlines, particularly those in rented accommodation. The survey is thus biased towards older people.

Nye Cominetti, an economist with the Resolution Foundation think tank, points out that the survey, which dates back 50 years, had a working-age sample of 99,000 in 1992, but just 31,000 in the latest quarter. For some groups in that sample, the numbers are not large enough to be accurate.

You may ask why this matters. The answer is that accurate information on the labour market matters a lot for policy, and for the interest ratesetters at the Bank of England, where it has become a key indicator.

Even on the most basic question of how many people are in work, the LFS throws up a conundrum. The latest official figures for employment based on it a month ago showed that total employment in May to July this year, just under 33 million, was fractionally lower than at the end of 2019, before the pandemic. Two other employment measures, however, show a very different picture. The ONS’s workforce jobs measure, largely based on a survey of employers, showed 36.7 million people in work in June, a rise of a million since December 2019. So-called real-time HMRC data, covering employees only, so not the self-employed, also showed a rise of more than a million compared with late 2019 levels, to 30.1 million.

Covid forced the ONS to rely on phone polling but many people no longer use landlines
Covid forced the ONS to rely on phone polling but many people no longer use landlines
GETTY IMAGES

These are big differences, with significant implications. If there has been a big rise in employment since before Covid, simple arithmetic would suggest there has been a sizeable drop in productivity. There are other implications. The ONS publishes figures for numbers of EU and non-EU workers in the UK but says these should be “used with caution” because they are based on 2021 patterns of migration. A plan to reweight the LFS this month has now been put off until March next year.

This is not the only problem faced by the ONS. There are longstanding issues with the reliability of migration statistics. Separately, it stunned economists and inadvertently launched a thousand conspiracy theories when it released revised GDP data at the start of last month, lifting the UK from worst G7 performer during the pandemic to mid-table. The Office for Statistics Regulation is conducting a review, though it was asked to do so by the ONS, which believes such revisions are a fact of life for GDP data and that there’s nothing much to see here.

Some people attribute the ONS’s problems to its shift from London to Newport in 2006-07. The move cost the ONS hundreds of experienced staff who decided that the other side of the Severn Bridge was not for them. But that was a long time ago and, pun intended, it is now water under the bridge. Newport, which has become something of a base for commuters to Cardiff and Bristol, has featured in a couple of recent surveys as the second and third most desirable place to live in the UK respectively, which surprised me. But there is still an issue of having the country’s statistical agency so far from the seat of national government.

As for the reliability of economic statistics, it is only fair to point out that the pandemic posed intense problems, many of which have also been experienced by statistical agencies of other nations. It is also fair to say that the ONS was innovative in its use of new data sources during Covid, and that its infection survey, now discontinued, provided the most accurate information on the spread of the disease.

We should not be too gloomy. Economic statistics are moving from the era of the clipboard and face-to-face interview to big data which, privacy concerns aside, should provide more complete information. Statistics should become better, not worse, though the ride is proving to be a bumpy one.

David Smith is Economics Editor of The Sunday Times
david.smith@sunday-times.co.uk

Sunday, 4 December 2022

The problem with statistics

 Call me geeky but I found this quite fascinating - errors in census data can have a big impact on funding for schools and more. It is one of those things it is useful to have in your toolkit in case you are making a point about policy directed at very specific issues and want to question whether the data the decision is based on is reliable:


Can we trust what the census says?

Crucial policy decisions are based on its data, but it’s only as accurate as the people filling it in: us

Tom Calver
The Sunday Times

When the 2011 census said just 9,000 people lived in Whitechapel, east London, Dr Shlomit Flint Ashery refused to believe it.

Armed with a clipboard and a grasp of Arabic and Bengali, the researcher at Bar Ilan University, Israel, conducted her own door-to-door survey of the ethnically diverse area. She befriended religious leaders in mosques and mothers in local parks. After weeks of work she had recorded the names of 13,000 individuals – 4,000 more than the census thought.

“The census is generally intended for the mainstream population,” says Flint Ashery, “and it tends to be blind to people from other cultures.” Specifically, in Whitechapel, it was not very good at counting households that contained multiple generations, or multiple families – especially among tight-knit migrant communities where distrust of the authorities is high.

Unlike the biblical iteration that required Joseph of Nazareth to return to his hometown, the modern census is a “snapshot” of the day it is filled in
Unlike the biblical iteration that required Joseph of Nazareth to return to his hometown, the modern census is a “snapshot” of the day it is filled in
ALAMY

The census is hailed as the gold standard: the Office for National Statistics (ONS) invests millions in ensuring it is done correctly. Filling it in is mandatory — those who didn’t in 2021 faced a £1,000 fine — and last time 89 per cent did so online. In Scotland, a dismal response rate of 79 per cent has forced an inquiry. Everything from homes, healthcare, schools and even sewer capacity is planned from the information it collects. Can we trust it?

In 1911 the suffragette Emily Davison, later killed by the King’s horse at Epsom, was found “hiding in the crypt of Westminster Hall” by census officials. This was recorded as her address.

Here lies a limitation of the modern census: unlike the biblical iteration that required Joseph of Nazareth to return to his hometown, it is a “snapshot” of the population of where they are on the day it is filled in.

The latest census was conducted in early 2021. Britain was in lockdown, so many of us were not where we usually live. Students who had gone home for Christmas did not immediately return to university; families may have gone to live with grandparents or in second homes.

Officials believe this made the population of some London boroughs seem much less than it usually is. Camden’s population — which the ONS previously thought was rising — nominally shrank by 10,000 over ten years to 210,000. Councillors, fearing the borough could suffer a reduction in government funding as a result, have joined a campaign to find the city’s “lost Londoners”.

The apparent drops are most significant among children, which could have implications for school funding. “Across London, billions of pounds for frontline services are at stake, and our residents will be the ones who suffer,” the Camden council leader Georgia Gould said in the summer.

Perhaps a greater problem is that the census is simply not very good at reaching certain communities. Some 97 per cent of households responded to the census, according to the ONS, up from 94 per cent a decade earlier: Professor Tony Champion, emeritus professor of population geography at Newcastle University, points out that the fact it happened “when most people were bored out of their minds and had plenty of time to complete the form” may have helped.

There are other reasons to treat census figures with a pinch of salt, as another population-wide programme conducted at the same time shows. Some 6.1 million people in their sixties had received a first vaccine dose by the end of 2021 — around 1 per cent more than the total number of sixty-somethings recorded in the census.

The latest census was conducted in early 2021 and recorded the population of England and Wales as 59,597,300
The latest census was conducted in early 2021 and recorded the population of England and Wales as 59,597,300
MATT ALEXANDER/PA

Comprehension has mercifully improved since the first census forms were handed out 200 years ago. When asked of his “relationship” to the head of the household, one Victorian agricultural labour, living in a shed at the bottom of a farmer’s garden, replied “friendly”. Yet even now that most of us can read, some of the categories remain ambiguous. If you are not one of Britain’s two million churchgoers but enjoy some fruits of the Christian tradition, do you mark your religion as Christian? Even ethnicity is, apparently, not that objective. “I changed my ethnicity between 2011 and 2021,” says Professor Edward Higgs, a leading census historian: “I put white Irish this time.”

Censuses are good at tracking change — but not if you change the questions. For last year’s census, the ONS moved “British” to the top of the census form, where “English” had been a decade before.

This was, it says, to make it easier to answer the question — yet in a striking example of “question-order bias”, the number of people identifying as “English” fell from 58 per cent, to 15 per cent. Some of that fall may be a genuine collapse in “English” identity, but we have no way of knowing.

Does any of this matter? It very much can do. For years, the ONS forecast that Coventry’s population would rise by 32 per cent between 2011 and 2031 – twice as fast as Birmingham’s. Yet on closer inspection, it emerged that these projections expected most of the city’s substantial student population to stay in Coventry after graduating (most do not). Thousands of homes for these graduate ghosts that were set to be built — some on greenbelt land — were not really needed.

There is a solution to these census imperfections. Countries like Norway have a joined-up “population register”, where data on taxation, benefits, schooling, health and education — currently separate in Britain — are all linked together in a single database. This approach may not sit well with some privacy campaigners, and the ONS is set to decide next year whether it can adopt a similar approach here.

This most recent census may well end up being the “best and highest quality” so far, according to Champion. Yet as long as it is filled in manually, it can only ever be as accurate as the people filling it in.

@TomHCalver